Background: HIVNET 012 was a phase II randomized open label trial in Uganda which demonstrated the overall efficacy of  the two dose NVP given to HIV infected mothers at labor onset and to their neonates compared to intrapartum and 1 week post partum zidovudine (ZDV) for reducing perinatal HIV transmission. Subgroup analyses of 18 month transmission rates  by maternal viral load and CD₄  count are described below.

Methods:  Subgroup analyses  by viral load RNA copy/ml (VL) and CD₄ were done using Kaplan Meier survival analyses. VL cutpoints  were <500, 500-9999, 10,000-49,999,>=50,000; and CD₄  were <200, 200-349; 350-499; >= 500. Transmission outcomes were based on infant RNA PCR.

Results: Subgroup data for 583 women in the trial were analyzed.  Lower 18 mo transmission rates for women on NVP compared to ZDV were noted across almost all  strata but differences were greatest among women with the highest VL and lowest CD_4.  Relative transmission risk: 1.9 (1.2-3.3) for women on ZDV vs NVP with VL >50,000.

    18 month transmission Rates (%) by VL and CD4 Subgroups, KM Method

VL RNA copies/ml      <500        500-9999              10,000-49,999     >=50,000                

ZDV                                 0%          12.3%                   21.2%                44.6%

NVP                                 0%           6.2%                    17.3%               24.9%

CD₄  Counts                           >=500       350-499         200-349             <=200_____         

ZDV                                           12.4%        24.7%            29.8%              54.9%

NVP                                           11.2%        10.5%            22.6%              31.6%

Conclusions: Transmission was significantly lower among Ugandan women receiving NVP compared to ZDV overall and for those with the most severe immune suppression and highest viral burden.  These data support the efficacy of the simple NVP two dose peripartum regimen even among women with the most advanced HIV disease.